Claude Sonnet 4.5 vs GPT-5.5 Instant
Wins 2 of 8 areas
Agents · Long documents
Wins 6 of 8 areas
Coding · Reasoning · Facts · Images and charts · Math · Following instructions
GPT-5.5 Instant is the stronger all-rounder.Claude Sonnet 4.5 is cheaper and better at agents.
Scores updated · 26 tests both models report · How we compare
Where each one wins
Tests won in each of the eight areas we test. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
- ReasoningHard problems that need careful thinking03GPT-5.5 Instant3 of 3 tests
- Images and chartsUnderstanding pictures, charts and video02GPT-5.5 Instant2 of 2 tests
- CodingWriting and fixing software12GPT-5.5 Instant2 of 3 tests
- FactsGetting facts right instead of making them up12GPT-5.5 Instant2 of 3 tests
- MathCompetition and research-level math01GPT-5.5 Instant1 of 2 tests · 1 tie
- Following instructionsDoing exactly what it is asked01GPT-5.5 Instant1 of 1 test
- AgentsCarrying out multi-step tasks on its own21Claude Sonnet 4.52 of 3 tests
- Long documentsFinding answers in very long texts10Claude Sonnet 4.51 of 1 test
Long documents and following instructions rest on a single test each.
What it costs
Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.
Claude Sonnet 4.5 costs 49% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where GPT-5.5 Instant pulls ahead
- Reasoning about charts from research papersCharXiv (RQ)+14.4points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+14.2points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+13.4points ahead
Where Claude Sonnet 4.5 pulls ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+21points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+16.2points ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+12.7points ahead
Every test, side by side
All 26 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGPT-5.5 Instant
- Terminal-Bench 2.1Claude Sonnet 4.5 by 2155.834.8+21
- SciCodeGPT-5.5 Instant by 6.845.752.5+6.8
- Terminal-Bench HardGPT-5.5 Instant by 6.835.642.4+6.8
AgentsClaude Sonnet 4.5
- GDPValClaude Sonnet 4.5 by 20.520.50+20.5
- τ-Bench V3 · BankingClaude Sonnet 4.5 by 12.724.511.8+12.7
- Terminal-Bench 4.0GPT-5.5 Instant by 12.6012.6+12.6
ReasoningGPT-5.5 Instant
- Humanity's Last ExamGPT-5.5 Instant by 3.817.821.6+3.8
- CritPtGPT-5.5 Instant by 1.51.12.6+1.5
- GPQA DiamondGPT-5.5 Instant by 1.283.484.6+1.2
FactsGPT-5.5 Instant
- AA-Omniscience · Non-hallucinationClaude Sonnet 4.5 by 16.250.934.7+16.2
- AA-Omniscience · AccuracyGPT-5.5 Instant by 13.432.946.3+13.4
- SimpleQA VerifiedGPT-5.5 Instant by 7.330.738+7.3
Images and chartsGPT-5.5 Instant
- CharXiv (RQ)GPT-5.5 Instant by 14.467.281.6+14.4
- MMMU-ProGPT-5.5 Instant by 7.368.776+7.3
MathGPT-5.5 Instant
- FrontierMath Tiers 1-3 (v2)GPT-5.5 Instant by 2.423.926.3+2.4
- FrontierMath Tier 4tie2.42.4tie
Following instructionsGPT-5.5 Instant
- IFBenchGPT-5.5 Instant by 14.257.371.5+14.2
Other results8 tests, not counted
Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.
- τ²-Bench Telecom (AA run)Claude Sonnet 4.5 by 28.778.149.4+28.7
- Artificial Analysis Coding IndexClaude Sonnet 4.5 by 12.752.139.4+12.7
- AA-OmniscienceGPT-5.5 Instant by 11.3-0.111.2+11.3
- HealthBenchGPT-5.5 Instant by 7.244.251.4+7.2
- AIME 2025Claude Sonnet 4.5 by 5.88781.2+5.8
- AA IntelligenceGPT-5.5 Instant by 220.722.7+2
- AA Agentic IndexGPT-5.5 Instant by 1.717.519.2+1.7
- OmniDocBench 1.5GPT-5.5 Instant by 1.785.887.5+1.7
Questions people ask
Which is better, Claude Sonnet 4.5 or GPT-5.5 Instant?
GPT-5.5 Instant wins six of the eight areas we test: coding, reasoning, facts, images and charts, math and following instructions. Claude Sonnet 4.5 wins agents and long documents, and costs 49% less.
Which is better for coding?
GPT-5.5 Instant. It wins 2 of the 3 coding tests both models report; Claude Sonnet 4.5 wins 1.
Which is cheaper?
Claude Sonnet 4.5 costs $3.00 per million input tokens and $15.00 per million output tokens; GPT-5.5 Instant costs $5.00 and $30.00. That makes Claude Sonnet 4.5 about 49% cheaper for the same work.
How do you compare the two?
We use the 26 benchmark tests both models have published scores on. The verdict counts the 18 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 8 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.